first demonstration with beam of the achromatic telescopic ... · • beam1 captured and...
TRANSCRIPT
S. Fartoukh Perf. Workshop, Chamonix 2012 1
First demonstration with beam of the
Achromatic Telescopic Squeeze (ATS)
S. Fartoukh
The Achromatic Telescopic Squeeze (ATS): motivations and basic principles
2011 ATS MDs: Overview, Highlights and Lessons
Plans for 2012
Do we need the ATS for the LHC-2012 (pre-squeezed optics)?
Summary
Main References:
sLHC-PR0038 & sLHC-PR0049/0053 Phase I LHC Optics limitations & description of the ATS scheme
ATS Notes 2011-033, 2011-060, 2011-132 2011 ATS MD results
ATS optics repository for the existing LHC:
/afs/cern.ch/eng/lhc/optics/ATS_V6.503 ATS optics compatible with the layout version V6.503 of the LHC
Motivations
… Lower b* need magnets of larger aperture, but also new hardware or
sophistication (crab-cavity, flat optics …) to “profit from the low b*”
the HL-LHC Project
… But this does not tell how to produce the b* the ATS scheme
which solves many optics limitations coming from the overall ring:
1. Optics matchability to the arcs:
Some IR quads going to 0 T/m, others to max. field (200T/m)
Simply the matching section becomes too short at some point!
2. Correctability of the chromatic aberrations induced, not only
Q’, but also Q’’, Q’’’,…, and off-momentum b-beating.
Arc sextupole strength (<600A)
S. Fartoukh Perf. Workshop, Chamonix 2012 2
Phase I optics limitations
(Chamonix 2010)
The basic principles of the ATS
• A new injection optics (~ p/2 FODO lattice new integer tunes)
• A squeeze in 2 steps
1) An “almost” standard squeeze, the Pre-squeeze:
acting on the matching quads of IR1 and IR5,
with new matching constraints on the left/right IR phase
till reaching some limits (sextupoles, matching section).
2) A further reduction of b*, the Squeeze:
acting on IR2/8 for squeezing IR1 and IR4/6 for IR5,
inducing b-beating bumps in sectors 81/12/45/56 to boost
the sextupole efficiency at constant strength.
Perf. Workshop, Chamonix 2012 3
Mismatched
FODO
Arc
Arc*
SqueezePre
*
Squeeze ˆ
ˆ
b
bbb
S. Fartoukh
The ATS scheme
(Chamonix 2011)
4 Tunes vs. d p
IR5 aperture (compared to
4 sigma's (X-angle OFF)
Reaching b*=10 cm in the existing LHC for ATLAS & CMS … a small animation
S. Fartoukh Perf. Workshop, Chamonix 2012
/afs/cern.ch/eng/lhc/optics/ATS_V6.503
Beam size [mm] and dispersion (IR4IR6)
at 3.5 TeV (for ge=3.5 mm)
IR4 IR5 IR6
Chromatic Montague functions over the
full ring (off-momentum b -beating)
Highlights of the ATS MD over 2011
A full program deployed in 2011, with a total of 8h
spent for dry runs (w/o beam) + 23 h with beam:
1. Bloc 1 (2h+8h): ATS injection optics and ramp (new integer tunes,
new arc and IR optics).
successful
2. Bloc 2 (3h+9.5h): Pre-squeeze to b*=1.2 m in IR1&5 (no X-angle),
and telescopic squeeze demonstrated for IR1, reaching b*= 30 cm.
successful
3. Bloc 4 (2h+4.5h): Pre-squeeze to b*=40 cm in IR1&5 (no X-angle),
and squeeze of IR1 and IR5 down to b*= 10 cm.
Pre-squeeze successful, beam lost when preparing the squeeze (“misunderstanding” between QFB and new tune knobs)
S. Fartoukh Perf. Workshop, Chamonix 2012 5
• Beam1 captured and circulating at the first injection!
Perf. Workshop, Chamonix 2012 6
… using nominal injection MCB pre-settings, which means that the CO correction of the nominal optics is very local.
• 30 min. later: 0.15 mm r.m.s. closed orbit (crossing knobs off)
S. Fartoukh Thanks to J. Wenninger et al.
Highlights of the ATS MD Part I (new inj. optics and ramp)
• First tune measurement:
Machine fully coupled but with a DQ min of only 0.06 w/o ANY correction,
and then easily decoupled with the closest tune approach (down to 10-3 or less)
Perf. Workshop, Chamonix 2012 7
|C-| is given to be 0 in FPGA
while the machine is fully coupled
S. Fartoukh
Perf. Workshop, Chamonix 2012 8
New settings for the damper (phases) and
< 50 turns damping time at the first attempt
First dump test … with the beam right
onto the target
• Tests of LHC sub-systems at 450 GeV: damper, dump, RF trims
S. Fartoukh
Thanks to W. Hofle et al.
Thanks to B. Goddard et al.
Perf. Workshop, Chamonix 2012 9
• The new integer tunes
New integer tunes 62/60 (instead of 64/59)
S. Fartoukh
Perf. Workshop, Chamonix 2012 10
…The ramp: excellent life time, transmission of intensity, and no
emittance growth
S. Fartoukh
Perf. Workshop, Chamonix 2012 11
• Beta-beating measurements at 450 GeV and 3.5 TeV (inj. optics)
Beta-beating at 450 GeV (no correction)
30-35% dominated by the IRs (accuracy of the magnet model at low current)
Beta-beating at 3.5 TeV(no correction)
10-15% residual from the arcs (random b2)
Signature of the MQ sorting
S. Fartoukh
Thanks to the OMC team:
R. Miyamoto, R. Tomas, G. Vanbavinckhove et al.
Perf. Workshop, Chamonix 2012 12 S. Fartoukh
Lessons from the ATS MD Part I
The LHC performs as MAD, if not better
Extremely well optimized such that e.g. the
choice of integer tunes is transparent, the
fractional tunes can be widely moved for a
closest tune approach
Extremely well sorted at installation
A priori fully debugged
.. And much faster for simulations!
• Chromatic measurements of the pre-squeezed optics (b*=1.2 m).
Perf. Workshop, Chamonix 2012 13
The Tunes are linear vs. dp over a
momentum window of +/- 1.5 permil
The off-momentum b-beating wave
induced by the IT starts to show up
but is contained in s81/12/45/56
S. Fartoukh
Thanks to the OMC team:
R. Miyamoto, R. Tomas, G. Vanbavinckhove et al.
Highlights of the ATS MD Part II
(pre-squeeze to 1.2 m in IR1 and IR5, and telescopic squeeze to 30 cm at IP1)
S. Fartoukh Perf. Workshop, Chamonix 2012 14
• Telescopic principle (×4) demonstrated in IR1
Perf. Workshop, Chamonix 2012 15
Could have been much worst with a bmax of 7 km in IR1 and by increasing by a
factor of 4 the b’s in s81 and s12.
Optimization still needed (IR8, IR4, arc by arc fine-tuning of the b2 correction)
S. Fartoukh
Thanks to the OMC team:
R. Miyamoto, R. Tomas, G. Vanbavinckhove et al.
• Beta-beating measurements at b*= 4.4, 1.2, 0.54 m, and 30 cm (3 trims, extracted from the nominal optics and incorporated for Q2.R1, Q2.L5 and Q2.R5)
• Losses (hystory over the last 3 hours) with TCPH/V set at ~9 s in IR7,
TCTH/V set a 12/10 mm in IR1 and IR5.
Perf. Workshop, Chamonix 2012 16
AC-dipole at 1.2 m
AC-dipole at 1.2 m
with 120 HZ RF trims
AC-dipole at 54cm
AC-dipole at 30 cm
No losses except during AC-dipole measurement
S. Fartoukh
Courtesy of D. Wollmann
Perf. Workshop, Chamonix 2012 17 S. Fartoukh
Lessons from the ATS MD Part II
The “Telescopic” part of the ATS is demonstrated,
… at least for IR1.
Additional squeeze lever arm by a factor of 4!
The “Achromatic” part of the ATS is validated … at
least down to the “relaxed” pre-squeezed b* of 1.2 m.
Non-linear chroma and off-momentum b-
beating under control!
Perf. Workshop, Chamonix 2012 18
IR5 pre-squeeze at b*= 40 cm bmax= 6 km in the triplet
(s ~2.5 mm @3.5 TeV, ge = 3.5 mm)
FODO b’s in arcs 45 & 56
FODO dispersion
IR5 squeeze at b*= 10 cm bmax= 24 km in the triplet
(s ~5 mm @3.5 TeV, ge = 3.5 mm)
400% b-mismatch in arcs 45 & 56
FODO dispersion
Highlights of the ATS MD Part III
(pre-squeeze to 40 cm and telescopic squeeze to 10 cm both at IP1 & IP5)
S. Fartoukh
Perf. Workshop, Chamonix 2012 19
Only 25% of the RS circuits actively participates to the chromatic correction
and only during the pre-squeeze
b*< 40 cm During the telescopic part, the chromatic
correction is achieved by the b-beating! b*> 40 cm
• Pre-squeezed pushed to the limit for the sextupoles
S. Fartoukh
300 A is reached in some RSD circuits at 3.5 TeV,
which means 600 A at 7 TeV!
Perf. Workshop, Chamonix 2012 20
Only 20 mV margin left at the end
of the pre-squeeze (b*=40 cm)
~ 330 A for some Q6’s in IR1/5
@ b*=40 cm and 3.5 TeV
(would be 240A for the nominal collision
optics @ b*=60cm and 4.0 TeV)
• Pre-squeeze pushed to the limit for the RQ6 circuits of IR1 & IR5
Already low current @ 7 TeV, therefore was even lower @ 3.5 TeV!
S. Fartoukh
Perf. Workshop, Chamonix 2012 21
• Excellent transmission during ramp and fast pre-squeeze
Only 2h for 40 cm
S. Fartoukh
As fast as 3km/h in terms of bmax, including a stop at 1 m for optics measurement!
Perf. Workshop, Chamonix 2012 22
• Rather constant Q’ during the pre-squeeze
• Coupling easily corrected with global knobs, but some trend visible
Fine tuning of the RQSX’s needed in IR5 and rebalancing the arc a2 correction
S. Fartoukh
Local coupling measurement at b*=40 cm (B1)
Thanks to the OMC team: R. Miyamoto,
R. Tomas, G. Vanbavinckhove et al.
Perf. Workshop, Chamonix 2012 23
15-20% level reached for the b-beating after one single correction iteration (only Beam1)
Qx(d ) at b *=40 cm
Qy(d ) at b *=40 cm
No (small) Q’’, Q’’’ …
Thanks to the OMC team:
R. Miyamoto, R. Tomas, G. Vanbavinckhove et al.
S. Fartoukh
• Optics measurements and correction at b*= 40 cm
Perf. Workshop, Chamonix 2012 24 S. Fartoukh
Lessons from the ATS MD Part III
• For the LHC
The nominal b* (55 cm) or below (40 cm) is
within reach (from the optics point of view)!
• For the ATS & HL-LHC
So far so good …but still a lot of work ahead to
guaranty the success of 10 cm (local coupling, b-
beating and phase-beating correction at 40 cm, tune knobs
& Q-feed-back,…)
Plans for 2012 • Prepare a clean 40 cm pre-squeezed optics (no X-angle)
measured and corrected for coupling, b-beating, …
• Prepare a safe 40 cm pre-squeezed optics (no X-angle)
using tight collimator settings and high brightness bunches
for an high pile up facility (up to ~120 events/crossing)
• Approach and hopefully reach & measure the 10 cm b*
with pilot beam.
• If time permits, flat pre-squeezed/squeezed optics
10-15% more performance after LS1 (from the lumi loss factor)
Plan B for an HL-LHC w/o crabs (but with 40% less performance)
S. Fartoukh Perf. Workshop, Chamonix 2012 25
Do we need the ATS (pre-squeeze) for 2012? Will depend on which limit will be met first at b*=60 cm
Triplet aperture or chromatic aberrations possibly
impacting on collimation, background, synchro-betatron
S. Fartoukh Perf. Workshop, Chamonix 2012 26
25-35% in the IT and collimation IR3 & IR7
i.e. ~ 10-15% level reached at the bucket separatrix
and much more for un-captured beam!
Only 1-2% in the IT and collimation IRs
Nominal optics @ b*= 60 cm
Db/b [%] at d=0.001 (Beam1, H plane)
ATS (pre-squeeze) @ b*= 60 cm
Db/b [%] at d=0.001 (Beam1, H plane)
ATLAS CMS
Other concerns: Tune vs. dp possibly impacting on
collective effects, sensitivity to RF trims, resonances,…
S. Fartoukh Perf. Workshop, Chamonix 2012 27
Nominal optics @ b*=60 cm ATS (pre-squeeze) @ b*=60 cm
Octupole OFF
450A in MO’s at 4 TeV 450A in MO’s at 4 TeV
Octupole OFF
Bucket ~ +/- 0.4 permil
Summary The aim of the ATS is twofold:
1. Produce ultra-low b* optics flat or round) for HL-LHC
2. With specific achromatic properties, also very attractive for
LHC (pre-squeezed optics down to 40 cm)
Beam validation of the ATS for round optics: 1. Several key milestones already reached, but separately
- Achromatic pre-squeeze pushed down to the limit (b* = 40 cm at IP1&IP5)
- Telescopic principle (×4) to reach b* = 30 cm in IR1 (starting from 1.2 m)
2. But some pieces still missing, everything to be combined at the end
- Telescopic squeeze of IR5 (using IR4 & IR6)
- Produce and check the properties of the fully squeezed optics (b*=10 cm)
S. Fartoukh Perf. Workshop, Chamonix 2012 28
S. Fartoukh Perf. Workshop, Chamonix 2012 29
Acknowledgments
… Many thanks
for these results which are the fruit of the work of many
people: the “pillars” of the ATS MDs, but also many others
C. Alabau, M. Albert, R. Alemany Fernandez, R. Assmann, R.
Bruce, A. Butterworth, R. Giachino, B. Goddard, P. Hagen, W.
Hofle, D. Jacquet, M. Giovannozzi, V. Kain, G. Kruk, M. Lamont,
E. Maclean, A. Macpherson, R. de Maria, R. Miyamoto, G.
Mueller, L. Normann, G. Papotti, M. Pojer, L. Ponce, S. Redaelli,
N. Ryckx, R. Steinhagen, M. Strzelczyk, R. Suykerbuyk, E. Todesco,
R. Tomas, D. Valuch, V. Venturini, G. Vanbavinckhove, J.
Wenninger, D. Wollmann, F. Zimmermann
Reserve
S. Fartoukh Perf. Workshop, Chamonix 2012 30
Why does it work? The magic lies in the choice of the betatron phases
.. Zoom in arc45: Pre-squeeze to 40 cm
Perf. Workshop, Chamonix 2012 31 S. Fartoukh
Dmy p between
strong SD’s (every other
FODO cells)
Dmy (Q11 IP ) = 1.25×p + ey
with ey ~ 1/2 tan-1 (bmin/bmax)
barc× b*)V cst
Dmx p between
strong SF’s (every other
FODO cells)
Dmx(Q14 IP ) = 1.25×p + ex
with ex ~ - 1/2 tan-1 (bmin/bmax)
barc× b*)H cst
p/2 phase in the arc cells
~ p/2 on the left & right side of the IR
2 sextupole families available per plane but
only one per plane used for triplet correction
Why does it work? The magic lies in the choice of the betatron phases
.. Zoom in arc45: Squeeze to a flat optics with b*=20cm/5cm at IP5
Dmy p between
strong SD’s (every other
FODO cells)
Dmy (Q11 IP ) = 1.25×p + ey
with ey ~ 1/2 tan-1 (bmin/bmax)
barc× b*)V cst
Dmx p between
strong SF’s (every other
FODO cells)
Dmx(Q14 IP ) = 1.25×p + ex
with ex ~ - 1/2 tan-1 (bmin/bmax)
barc× b*)H cst
Perf. Workshop, Chamonix 2012 32 S. Fartoukh
ATS-MD-II: coupling measurement at b*= 4.4, 1.2, 0.54 and 30
cm (with RQSX pre-setting from nominal optics implemented at 450GeV)
Perf. Workshop, Chamonix 2012 33
No time to correct the coupling below 1.2 m but we were prudent enough to work
with the injection tune, i.e. a tune split of 0.03 (see later the QFB history).
C-
C+
S. Fartoukh
Thanks to the OMC team:
R. Miyamoto, R. Tomas, G. Vanbavinckhove et al.
ATS-MD-II: Feed-backs (trim hystory over the last hours)
Perf. Workshop, Chamonix 2012 34
Start the squeeze
below 1.2 m
Some correction and
incorporation at 54 cm
30 cm!
Nominal tune jump compensated on purpose by
the QFB (preventive tune split for coupling)
Rather smooth for the FIRST squeeze of a “8km long insertion” (3 IRs+2 arcs)!
S. Fartoukh
Orbit
Tunes
Perf. Workshop, Chamonix 2012 35
ATS-MD-I: Dispersion measurements at 450 GeV and 3.5 TeV
Spurious Dispersion at 450 GeV
50-60 cm in the arcs in the H plane
15-20 cm in the arcs in the V plane
Spurious Dispersion at 3.5 TeV
No improvement!!
- Dominated by the contribution from the arcs (no change from 450 GeV to 3.5 TeV, contrary to b-beating)
- Small in the V plane: Signature of the MB sorting on the random a2
- Sizeable in the H plane: MQ sorting only optimized for beta-beating (too many SSS types)
…
Will increase with sqrt(bmax) in the IT during the squeeze !
S. Fartoukh
ATS-MD-II: Dispersion measurements at b*= 4.4 m, 1.2 m and 54
cm. The “worst” case of Beam2 at 54 cm is shown.
Perf. Workshop, Chamonix 2012 36
The spurious dispersion from the arcs shows up clearly in the inner triplet
(depending on whether we are lucky or not with the betatron phases)
IR1: 1m at 4.5 km b max
IR1:60 cm at 4.5km b max
IR5:1m at ~2km b max
S. Fartoukh
Perf. Workshop, Chamonix 2012 37
Dy still acceptable
Spurious Dx starts to be worrying, but a priori correctable
80 cm at 6km b max
1m at 6km b max
1.6m at 6km b max
ATS-MD-III: Dispersion measurements at b*= 40 cm (the worst case of
Beam2 is shown)
S. Fartoukh
Perf. Workshop, Chamonix 2012 38
Beam1 Beam2
Triplet clearly tilted in IR5 while IR1 looks OK … but at 10 cm who knows?
RQSX fine tuning needed (local correction) and global correction to be re-
worked during the pre-squeeze sequence.
• Local coupling measurements at b*= 40 cm
S. Fartoukh
Thanks to the OMC team:
R. Miyamoto, R. Tomas, G. Vanbavinckhove et al.
Perf. Workshop, Chamonix 2012 39 S. Fartoukh
ATS-MD-III: Beam lost when preparing the telescopic part of the squeeze
Real time trims sent to zero by the QFB when switching between the two tune knobs to be used above and below the pre-squeezed b*
Several RQT circuits tripped by QPS
Not a conceptual limit, with some idea already proposed
Courtesy of M. Pojer and R. Steinhagen
Other possible developments
• Rework a bit the ATS to transform it into a Telescopic Un-
squeezing scheme for very large b* (forward physics)
• Use the ATS as IBS knob in the longitudinal plane Possible application for Heavy Ions but the corresponding 1m ATS optics in
IR1/5 and IR2 is not ready (even not tried on paper)
• Any other “exotic’’ application: much more Landau
damping available for impedance, or even shaping the
head-on beam-beam tune spread, …
S. Fartoukh Perf. Workshop, Chamonix 2012 40